Active Learning Driven Data Acquisition for Sensor Networks

Abstract

Online monitoring of a physical phenomenon over a geographical area is a popular application of sensor networks. Networks representative of this class of applications are typically operated in one of two modes, viz. an always-on mode where every sensor reading is streamed to a base station, possibly after in-network aggregation, and a snapshot mode where a user queries the network for an instantaneous summary of the observed field. However, a continuum of data acquisition policies exists between these two extreme modes, depending upon the rate and manner in which each sensor node is queried. In this work, we explore this continuum to improve network energy efficiency.We present a data acquisition framework that models the evolution of the observed data field at each sensor location as a function of time and uses an active learning based criterion to intelligently sample each sensor. Sensor nodes in our framework are organized in a clustered hierarchy. Time-dependent models of sensor readings are maintained at cluster-head nodes, which sample nodes in their cluster in a way that minimizes total energy consumption while maintaining confidence bounds on the overall model. We use sparse Gaussian processes to model sensor readings and variance minimization based active learning to intelligently select sensor nodes for querying. Finally, we present simulation results demonstrating up to 70% savings in total network energy, compared to the base case, in which sensors are sampled according to a cyclic schedule.

Energy efficient information collection in wireless sensor networks using adaptive compressive sensing

“…This work distinguishes itself from other works in energy efficient adaptive sensing, e.g. [6], [11], [13], [16], in that it uses recent works in adaptive compressive sensing [10], [20]. However, existing work in adaptive compressive sensing only takes into consideration the accuracy of the approximate data field.…”

“…We generate correlated data on the sensor network using an algorithm similar to the one used in [16]. Let d ij be the distance between sensors i and j, then we assume that the correlation of the data between them is given by exp(−0.5d ij ).…”

“…[6], [11], [13], [16]. Our work distinguishes from these earlier works in that it uses adaptive compressive sensing.…”

“…Although there is a rich literature on adaptive sampling for WSNs, e.g. [6], [11], [13], [16], to the best of our knowledge, our algorithm is the first one that uses adaptive compressive sensing.…”

Automated Assessment of Aggregate Query Imprecision in Dynamic Environments

“…Researchers in sensor networks have explored a model-driven approach to query processing [14,15]. Each node constructs a local model of the data in the network and estimates the error in the model.…”